检索结果-南通市图书馆

MODELING THE knowledge CREATION PROCESS OF GRADUATE RESEARCH

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Systems science and Systems Engineering 2018年第6期27卷 790-809页

作者： J. Sun J. Tian V.N. Huynh Y. Nakamori school of knowledge science Japan Advanced lnstitute of Science and Technology lshikawa Japan school of Computer science and Technology Wuhan University of Technology Wuhan China

Graduate students who are beginning academic research want to learn how to create and verify new knowledge in their research. Their supervisors, on the other hand, are seeking appropriate research environments including effective research guidance methods. In order to meet these demands, this paper proposes a knowledge creation model that supports the objectives of both graduate students and their supervisors. This is an academic knowledge creation model for individuals supported by a group and its origin can be traced back to a famous organizational knowledge creation model. Since this type of model is constructed from empirical knowledge, it is not easy to prove its objective significance. But, this paper tries to show the effectiveness of the proposed model as an initial stage of model validation based on a questionnaire survey of students in a graduate school in China.

关键词： knowledge creation model graduate students research ability research environment

Reasoning and predicting POMDP planning complexity via covering numbers

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer science 2016年第4期10卷 726-740页

作者： Zongzhang ZHANG Qiming FU Xiaofang ZHANG Quan LIU school of Computer science and Technology Soochow University Suzhou 215000 China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing 210000 China school of Electronic and Information Engineering Suzhou University of Science and Technology Suzhou 215000 China Key Laboratory of Symbolic Computation and knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

Partially observable Markov decision processes （POMDPs） provide a rich mathematical framework for planning tasks in partially observable stochastic environments. The notion of the covering number, a metric of capturing the search space size of a POMDP planning problem, has been proposed as a complexity measure of approximate POMDP planning. Existing theoretical results are based on POMDPs with finite and discrete state spaces and measured in the l1- metric space. When considering heuristics, they are assumed to be always admissible. This paper extends the theoretical results on the covering numbers of different search spaces, including the newly defined space reachable under inadmissible heuristics, to the ln-metric spaces. We provide a simple but scalable algorithm for estimating covering numbers. Experimentally, we provide estimated covering numbers of the search spaces reachable by following different policies on several benchmark problems, and analyze their abilities to predict the runtime of POMDP planning algorithms.

关键词： POMDP planning complexity theory covering number search space

A parallel scheduling algorithm for reinforcement learning in large state space

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer science 2012年第6期6卷 631-646页

作者： Quan LIU Xudong YANG Ling JING Jin LI Jiao LI Institute of Computer science and Technology Soochow UniversitySuzhou 215006China Department of Computer science and Technology Nanjing UniversityNanjing 210093China Key Laboratory of Symbolic Computation and knowledge Engineering of Ministry of Education Jilin UniversityChangchun 130012China

The main challenge in the area of reinforcement learning is scaling up to larger and more complex problems. Aiming at the scaling problem of reinforcement learning, a scalable reinforcement learning method, DCS-SRL, is proposed on the basis of divide-and-conquer strategy, and its convergence is proved. In this method, the learning problem in large state space or continuous state space is decomposed into multiple smaller subproblems. Given a specific learning algorithm, each subproblem can be solved independently with limited available resources. In the end, component solutions can be recombined to obtain the desired result. To ad- dress the question of prioritizing subproblems in the scheduler, a weighted priority scheduling algorithm is proposed. This scheduling algorithm ensures that computation is focused on regions of the problem space which are expected to be maximally productive. To expedite the learning process, a new parallel method, called DCS-SPRL, is derived from combining DCS-SRL with a parallel scheduling architecture. In the DCS-SPRL method, the subproblems will be distributed among processors that have the capacity to work in parallel. The experimental results show that learning based on DCS-SPRL has fast convergence speed and good scalability.

关键词： divide-and-conquer strategy parallel schedule scalability large state space continuous state space

Efficient reinforcement learning in continuous state and action spaces with Dyna and policy approximation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer science 2019年第1期13卷 106-126页

作者： Shan ZHONG Quan LIU Zongzhang ZHANG Qiming FU school of Computer science and Technology Soochow UniversitySuzhou 215000China school of Computer science and Engineering Changshu Institute of TechnologyChangshu 215500China Jiangsu Province Key Laboratory of Intelligent Building Energy Efficiency Suzhou University of Science and TechnologySuzhou 215006China Key Laboratory of Symbolic Computation and knowledge Engineering of Ministry of Education Jilin University Changchun 130012China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing 210000China College of Electronic &Information Engineering Suzhou University of Science and TechnologySuzhou 215000China

Dyna is an effective reinforcement learning (RL) approach that combines value function evaluation with model ***,existing works on Dyna mostly discuss only its efficiency in RL problems with discrete action *** paper proposes a novel Dyna variant,called Dyna-LSTD-PA,aiming to handle problems with continuous action ***-LSTD-PA stands for Dyna based on least-squares temporal difference (LSTD)and policy ***-LSTD-PA consists of two simultaneous,interacting *** learning process determines the probability distribution over action spaces using the Gaussian distribution;estimates the underlying value function,policy,and model by linear representation;and updates their parameter vectors online by LSTD(,t).The planning process updates the parameter vector of the value function again by using ofttine LSTD(2).Dyna-LSTD-PA also uses the Sherman-Morrison formula to improve the efficiency of LSTD(,t),and weights the parameter vector of the value function to bring the two processes ***,the global error bound is derived by considering approximation,estimation,and model ***,Dyna-LSTD-PA outperforms two representative methods in terms of convergence rate,success rate,and stability performance on four benchmark RL problems.

关键词： problem solving control methods heuristic search methods dynamic programming

Institutional Quality,Foreign Aid and Economic Growth in Recipient Countries

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Economist 2020年第6期15卷 68-83页

作者： Wang Xiaosong Tian Siyuan school of Economics Renmin University of China(RUC)BeijingChina

Foreign aid is an important aspect of China’s Belt and Road Initiative (BRI)for enhancing two-way cooperation with BRI countries. Based on the panel data of China’sforeign aid over the period 2000-2014, this paper employs the Worldwide GovernanceIndicators (WGI) to conduct an empirical study on the economic growth effects of China’sforeign aid and recipient countries’ institutional quality with the endogeneity of aid takeninto account. Results of our empirical study suggest that (i) China’s foreign aid, especiallyinfrastructure aid, can effectively promote economic growth in recipient countries;(ii)sound institutional development offers an important assurance for economic growth inrecipient countries;(iii) sample-specific regression reveals that the institutional quality ofrecipient countries can significantly influence the economic growth effects of China’s foreignaid, especially economic infrastructure aid for recipient countries in Asia and *** improve aid effectiveness and quality, China needs to improve aid structure, superviseaid program implementation, and mitigate the impact of institutional risks in recipientcountries.

关键词： foreign aid institutional quality economic growth BRI

Dicoumarol is an effective post-exposure prophylactic for SARS-CoV-2 Omicron infection in human airway epithelium

在线全文

学校读者我要写书评

暂无评论

Signal Transduction and Targeted Therapy 2023年第7期8卷 3455-3465页

作者： Yang Peng Shi-ying Chen Zhao-ni Wang Zi-qing Zhou Jing Sun Gui-an Zhang Jia Li Lei Wang Jin-cun Zhao Xiao Xiao Tang De-Yun Wang Nan-shan Zhong State Key Laboratory of Respiratory Disease National Clinical Research Center for Respiratory DiseaseGuangzhou Institute of Respiratory HealthThe First Affiliated Hospital of Guangzhou Medical UniversityGuangzhou Medical UniversityGuangzhouGuangdongChina Department of Otolaryngology Infectious Diseases Translational Research ProgramYong Loo Lin School of MedicineNational University of SingaporeSingaporeSingapore Guangzhou Laboratory GuangzhouChina

Repurposing existing drugs to inhibit SARS-CoV-2 infection in airway epithelial cells(AECs)is a quick way to find novel treatments for *** screening has found dicoumarol(DCM),a natural anticoagulant,to be a potential SARS-CoV-2 inhibitor,but its inhibitory effects and possible working mechanisms remain *** air-liquid interface culture of primary human AECs.

关键词： Omi infection epithelium

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

深度强化学习综述

计算机学报 2018年第1期41卷 1-27页

作者：刘全翟建伟章宗长钟珊周倩章鹏徐进苏州大学计算机科学与技术学院江苏苏州215006 软件新技术与产业化协同创新中心南京210000

深度强化学习是人工智能领域的一个新的研究热点.它以一种通用的形式将深度学习的感知能力与强化学习的决策能力相结合,并能够通过端对端的学习方式实现从原始输入到输出的直接控制.自提出以来,在许多需要感知高维度原始输入数据和决策... 详细信息

深度强化学习是人工智能领域的一个新的研究热点.它以一种通用的形式将深度学习的感知能力与强化学习的决策能力相结合,并能够通过端对端的学习方式实现从原始输入到输出的直接控制.自提出以来,在许多需要感知高维度原始输入数据和决策控制的任务中,深度强化学习方法已经取得了实质性的突破.该文首先阐述了三类主要的深度强化学习方法,包括基于值函数的深度强化学习、基于策略梯度的深度强化学习和基于搜索与监督的深度强化学习;其次对深度强化学习领域的一些前沿研究方向进行了综述,包括分层深度强化学习、多任务迁移深度强化学习、多智能体深度强化学习、基于记忆与推理的深度强化学习等.最后总结了深度强化学习在若干领域的成功应用和未来发展趋势.

关键词：人工智能深度学习强化学习深度强化学习

在线全文

学校读者我要写书评

暂无评论

2023年度全球工程前沿

Engineering 2023年第12期31卷 4-7页

作者： Jianlin Yan Zhirui Mu Wenjiang Zheng Center for Strategic Studies Chinese Academy of EngineeringBeijing 100088China

An engineering front is a key direction in engineering that is forward looking,cutting edge,and *** plays a leading role in the future development of engineering science and technology,serving as an important guide for innovation in engineering science and *** to its innovation stage,an engineering front is categorized into either an engineering research front,which focuses on theoretical exploration,or an engineering development front,which focuses on the practical application of engineering science and technology[1,2].

关键词： technology looking engineering

The effects of a 20-week exercise program on blood-circulating biomarkers related to brain health in overweight or obese children:The ActiveBrains project

在线全文

学校读者我要写书评

暂无评论

Journal of Sport and Health science 2023年第2期12卷 175-185页

作者： María Rodriguez-Ayllon Abel Plaza-Florido Andrea Mendez-Gutierrez Signe Altmäe Patricio Solis-Urra Concepción M.Aguilera Andrés Catena Francisco B.Ortega Irene Esteban-Cornejo PROFITH“PROmoting FITness and Health through physical activity”research group Sport and Health University Research Institute(iMUDS)Department of Physical and Sports EducationFaculty of Sport SciencesUniversity of GranadaGranada 18071Spain Department of Epidemiology Erasmus University Medical CenterRotterdam 3015 GDthe Netherlands Department of Biochemistry and Molecular Biology Ⅱ School of PharmacyUniversity of GranadaGranada 18071Spain Department of Biochemistry and Molecular Biology Faculty of SciencesUniversity of GranadaGranada 18071Spain Nuclear Medicine Services Virgen de Las Nieves University HospitalGranada 18014Spain Faculty of Education and Social sciences Universidad Andres BelloViña del Mar 2531015Chile Department of Experimental Psychology School of PsychologyUniversity of GranadaGranada 18011Spain Faculty of Sport and Health sciences University of JyväskyläJyväskylä40014Finland Department of Bioscience and Nutrition Karolinska InstitutetHuddingeSE 14157Sweden

Background:Emerging research supports the idea that exercise positively affects ***,the mechanisms linking exercise with brain health are largely *** aimed to investigate the effect of exercise on(a)blood biomarkers selected based on previous evidence(brainderived neurotrophic factor,β-hydroxybutyrate(BHB),cathepsin B(CTSB),kynurenine,fibroblast growth factor 21(FGF21),soluble vascular cell adhesion molecule-1(sVCAM-1));and(b)a panel of 92 neurology-related proteins(discovery analysis).We also investigated whether changes in these biomarkers mediate the effects of exercise on brain health(hippocampal structure and function,cognitive performance,and mental health).Methods:We randomized 81 overweight/obese children(10.1±1.1 years,41%girls)into 2 groups:either 20 weeks of aerobic plus resistance exercise or *** biomarkers were assessed using enzyme-linked immunosorbent assay(ELISA)for kynurenine,FGF21,and CTSB;colorimetry forβ-hydroxybutyrate;and XMap for brain-derived neurotrophic factor and soluble vascular cell adhesion ***92 neurology-related proteins were analyzed by an antibody-based proteomic ***:Our intervention had no significant effect on candidate biomarkers(all p>0.05).In the discovery analysis,a reduction in circulating macrophage scavenger receptor type-I was observed(standardized differences between groups=-0.3,p=0.001).This effect was validated using ELISA methods(standardized difference=-0.3,p=0.01).None of the biomarkers mediated the effects of exercise on brain ***:Our study does not support a chronic effect of exercise on candidate *** observed that while chronic exercise reduced the levels of macrophage scavenger receptor type-Ⅰ,it did not mediate the effects of exercise on brain *** studies should explore the implications of this novel biomarker for overall health.

关键词： Brain development Childhood MRI Physical activity Proteomic